HICKORY WOOD SMOKE
Tannenbaum, S.R., Barth, H., LeRoux, J. P., J , Agr. Food Chem. 17, 1353 (1969).
Tannenbaum, S.R., Bates, R. P.. Brodfeld, L., Food Technol. 24, 607 (1970).
Waterlow, J . C., Brit.J. Nutr. 16, 531 (1962). Wilson, R. F., Tilley, J. M. A,, J . Sci. Food Agr. 16, 173 (1965).
Woodham, A. A , , "Leaf Protein: Its Agronomy, Preparation, Quality and Use," Pirie, N. W., Ed., Blackwell Scientific Publications, London, 1971, pp 115-130. Yemm, E. W., Folkes, B. F., Biochem, J . 55,700 (1953). Received for review March 12, 1973. Accepted July 30, 1973.
Isolation and Identification of the Components of the Tar of Hickory Wood Smoke Denis E. Hruza, Sr.,* Michel van Praag,l and Howard Heinsohn, J r .
The volatile components of hickory wood tar were fractionated by preparative gas chromatography. Individual fractions were analyzed on a Carbowax 20M 50 ft x 0.02 in. support coated open tubular column (SCOT) coupled to a mass spectrometer. The fallowing compounds are among those being reported for the first time from hickory wood tar: 2-methyl-2-butenoic acid; benzaldehyde; 2-ace-
The smoking of foods as a method of preservation is one of the oldest methods known. Smoking not only partially dehydrates food during the process but also deposits compounds possessing antimicrobial antioxidant activities in the meat. With today's modern technology of food preservation, the smoking of foods is done for color and added flavor. Most of the previous work in the literature on smoke flavor has been carried out on wood smoke itself. The identification of acids, carbonyls, alcohols, and other neutral components has been published (Fiddler et al., 1967; Hamid and Saffie, 1965; Hoff and Kapsalopoulou, 1964; Jahnsen, 1961; Love and Bratzler, 1966; Porter et al., 1964). Recently, work was carried out on the constituents of a liquid smoke solution (Fiddler et al., 1970a,b). Much work has been done on the phenolic compounds and their role as contribut.ors to the flavor of smoke foods (Fiddler et al., 1967; Kornreich and Issenberg, 1972; Lustre and Issenberg, 1969; Tilgner et al., 1962). The technological aspects of the smoking process have revealed the presence of phenols in smoked foods. Since phenolic compounds identified in foods are not normal components, they were attributed to the smoking process (Fiddler et al., 1966; Foster and Simpson, 1961; Foster e t al., 1961; Porter et d., 1964; Tilgner l?t al., 1962; Ziemba, 1963). We decided to look a t a different aspect of smoke flavor by examining the tar of hickory wood smoke with a possibility of preparing a synthetic substitute. ISOLATIOS The material examined was a commerical sample obtained from Old Hickory Products Co., Atlanta, Ga. The material is a natural flavor product made from 100% hickory wood. As the hickory wood tar is a black viscous material, it was first dissolved in acetone (Matheson, Coleman and Bell, Spectroquality) and filtered to remove any carbon particles, i2nd most of the solvent was removed by distillation using ,a Kuderna-Danish concentrator (Kontes Glass Co., Vineland, N. J.). The clarified sample (ca. 8 ml) was then used for our analytical work. The acids present in the material were isolated and identified in the following manner. One kilogram of the hickory smoke concentrate was steam distilled and 2 1. of International Flavors and Fragrances, Inc., Union Beach, New Jersey 07735. International l?lavors and Fragrances, Tilburg, Holland.
tylfuran; 2-cyclohexenone; 4-propylguaiacol; 4methylveratrol; 3-methyl-2-cyclopenten-1-one; 2,3-pentanedione; acetophenone; resorcinol; vinylphenol; 2,6-dimethylphenol; and 2-ethylphenol. A total of six aldehydes, eight ketones, four esters, six furans, 12 aromatic hydrocarbons, 32 phenols, and 13 acids were identified.
distillate were collected. A liter portion was made basic to p H 7.8 with sodium bicarbonate and the nonacidic components were removed by extracting with 3 x 150 ml of diethyl ether (Matheson, Coleman and Bell, ACS reagent grade). The aqueous mixture was then made acidic with concentrated hydrochloric acid to p H 1, saturated with sodium chloride, and extracted with 3 X 150 ml of diethyl ether. The ethereal extracts were combined, dried over anhydrous sodium sulfate, and filtered, and the solvent was removed using a Kuderna-Danish concentrator. The residue was then methylated using a diazomethane generator utilizing N-methyl-Ai-nitroso-p-toluenesulfonamide (Aldrich Chemical Co.). The methylated residue was then analyzed in the same manner as the clarified sample described below. APU'ALYTICAL METHODS Preparative gas chromatography was carried out on an F&M 770 preparative gas chromatograph equipped with a thermal conductivity detector utilizing an 8 f t X 3/4 in. stainless steel column packed with 25% SE-30 on 60-80 mesh Chromosorb W A/W. The column temperature was programmed from 75" to 225" at 2"/min, with a helium carrier gas a t a flow rate of 300 ml/min. The injection sizes for this separation were 2 ml. Two milliliters of the clarified sample was injected and the effluent collected as 12 distinct peaks which were trapped in capillary glass tubes cooled with crushed Dry Ice and sealed (Figure 1). The fractions were further resolved using an F&M 700 glc equipped with a thermal conductivity detector utilizing in. stainless steel column packed with 20% an 8 f t X Carbowax 20M on 60-80 mesh Chromosorb W A/W. The column temperature was programmed from 75" to 225" a t 2"/min, with helium carrier gas a t a flow rate of 80 ml/ min. This column was used to trap out distinct peaks for nmr and ir analyses. Nmr spectra of the samples in deuteriochloroform solution were recorded using a Varian HA-100 spectrometer. Tetramethylsilane was the internal reference compound. Sweep width was 1000 Hz, with a 50-Hz sweep offset. Glc-ms analyses were carried out using an Aerograph 1520 gas chromatograph coupled to a Hitachi RMU-6E mass spectrometer. A 50 f t X 0.02 in. support coated open tubular (SCOT) stainless steel column coated with Carbowax 20M was used. The column temperature was programmed from 30" to 175" a t 2"/min, with a helium carrier gas a t a flow rate of 6 ml/min. The column was used with the column effluent split so that 5 ml was directed J . Agr.
Food Chem., Vol. 22, No. 1, 1974
123
HRUZA et al.
Table I. Compounds Identified from Tar of Hickory Wood Smoke Compound
Reported in smoke
Acids Propionic acid Butyric acid 2-Methylbutyric acid Isovaleric acid 2-Methylpropenoic acid Valeric acid 3-Methylpentanoic acid (T). 4-Methylpentanoic acid 2-Methyl-2-butenoic acid (cis) 2-Methyl-2-butenoic acid (trans) Hexanoic acid Octanoic acid Benzoic acid Aldehydes Acetaldehyde Propionaldehyde Tiglaldehyde (T) (2-Methyl-2-butenal) 2-Furfural Benzaldehyde 5-Methyl-2-furaldehyde Aromatic hydrocarbons Benzene Toluene Styrene (T) Indene Naphthalene 3-Methylindene 2-Methylnaphthalene Dimethylindane (T) Esters Methyl formate Methyl acrylate Ethyl benzoate Cresyl acetate Furans Furan 2-Methylfuran (T) 2-Ethylfuran (T) 2-Acetylfuran 2-Methylbenzofuran (T) Dimethylbenzofuran Ketones Acetone Methyl ethyl ketone 3- Methyl-3-butene-2- one (methyl isopropenyl ketone) 2,3-Butanedione (diacetyl) 2-Cyclohexenone 2,3-Pentanedione 3-Methyl-2-cyclopenten1-one Acetophenone Phenols Phenol o-Cresol m-Cresol p-Cresol 1,3-Dihydroxybenzene (resorcinol) Vinylphenol (o,m,p-?) 2,6-Dimethylphenol 2-Ethylphenol 2,4-Dimethylphenol 2,5-Dimethylphenol 2,3-Dimethylphenol 4-Ethylphenol 3-Ethylphenol o-Methoxyphenol (guaiacol) 124
J. Agr. Food Chern., Vol. 22,No. 1,1974
XC X
IE value, gc-ms
X
2.26 3.34 3.43 3.79
X
4.51
Mol wt
116 130 130 114
57-29-88-59-89-27 43-74-71-27-41-59-102 57-88-41-29-85-59-116 74-43-41-59-29-57-116 41-69-100-39-15-99 74-29-27-57-43-41-116 74-43-59-41-29-99-130 43-74-57-55-27-87-130 55-83-27-114-29-39
114
55-83-27-114-29-39 74-43-27-87-29-41-130 74-87-43-55-41-57-158 105-77-13651-50-29
88
102 116 116 100
4.63 5.59
X
Mass spectral data*
X X
5.53 7.56 10.00
130 158 136
x
0.44 0.81 4.61
44 58 84
29-44-43-26-42-27 29-28-27-58-26-57 55-29-84-27-39
8.24 9.03 9.36
96 106 110
39-96-95-66 106-77-105-57-50 110-109-53-27-51
X X
2.92 4.14 6.32 8.61 11.12 9.80 11.92
78 92 104 116 128 130 142 146
78-51-52-49-39-77 91-92-39-65-51 104-103-78-51-77 116-115-63-39-89 128-51-129-64-127-63 130-115-129-128-51-63 142-141-115-143-139 131-146-91-115-129-39
X
0.63 2.70 10.41 10.9
60 86 150 150
31-29-32-60-15 55-27-85-58-26-86 105-77-51-122-150 108-107-43-77-51-150
X X
0.82 2.00 3.00 8.70
82
X X X X
68 96 110 132 146
39-68-38-29-37 8.2-53-81- 39-27-29 81-39-96-41-51-65 95-110-39-43-96 131-132-51-77-39 146-145-117-39-115
X
1.00 2.21 2.70
58 72 84
43-58-27-26-42-29 43-29-27-72-42-57 43-41-39-84-69-42
X
3.17
86
43-86-42-44
8.15 4.13 8.73
96 100
110
68-39-96-40-27-42 43-29-57-29-100 82-39-110-54-27-41
10.26
120
105-77-120-51-43
13.38 13.08 13.98 14.11 10.50
94 108 108 108
94-39-66-65-40 108-107-79-77- 39 107-108-77-79 107-108-79-77 110-82-81-69-53-55
16.91 12.51 13.99 14.00 14.73 14.79 15.01 15.05 12.71
120 122 122 122 122 122 122
X X X X
x X
X X
110
122
12,4
120-91-26-119-39 122-107-121-77-39-91 107-122-77-39-79 122-107-121-77-39 122-107-121-77-39-27 107-122-121-77-39 107-122-77-39-27 107-122-77-39-27 109-124-81-53-27-39
HICKORY WOOD SMOKE
Table I (Continued) Compound
Rnported in smoke
Phenols m-Methoxyphenol Trimethylphenol (T) 3-E thyl-5-methylphenol
Mol wt
Mass spectral data* 124-94-81-39-53- 95 121-136135-39-91-77 121-13691-77-39-122 138-123-95-39-55-27
2-Methox:y-4-methylphenol (4-meth ylguaiacol)
X
12.98
2-Methoxy-4-vinylphenol (4-vinylguaiacol) tert-Butylphenol (T) 4-E thyl-2-methoxyphenol (4-ethylguaiacol) 2,6-Dimet hoxyphenol (syringcd) 4-Allyl-2-rnethoxyphenol (eugenol) 2-Methox!r-4-propenylphenol (isoeugenol) 4-Isopropyl-2-methoxyphenol (T) (4-isopropylgusiacol)
X
15.23
150
150-135-107-77-51-39
X
16.13 13.83
150 152
135-107-15&91-39-41 137-152-39-91-122-94
X
15.88
154
154-139-111-96-107-93
X
15.09
164
164-77-149-39-103
X
16.77
164
164-149-77-103-91-55-39
166
137-166-138-122-94-51
2-Methoxjr-4-propylphenol
X
14.41
166
137-166-138-122-94-51
X
16.77
168
168-153-125-53-65
X
17.26
182
167-182-77-107-79
X
18.99
194
194-91-119-77-167-39-131
X
21.18
194
194-91-77-119-79-39
196
181-196-137-77-91-121
196
167-196168-41-53-197
2.46 10.73
46 138
31-45-46-27-29-43 138-95-123-77-52-51
11.7 11.7
138 152
138-95-123-77-52-51 152-137-109-91-77-39
M:iscellaneous Ethanol 1,2-Dimethoxybenzene (veratro I) 1,4-Dimethoxybenzene (T) 1,2-Dimethoxy-4-methylbenzene f4-methylveratrol) a
gc-ms
124 136 136 138
(4-propjilguaiacol) 2,6-Dimetlhoxy-4-methylphenol (T) 2,6-Dimetlioxy-4-ethylphenol (T) 2,6-Dimetlioxy-4-allylphenol (T) 2,6-Dimetlioxy-4-propenyl phenol 2,6-Dimetlioxy-4isopropylphenol (T) 2,6-Dimetlioxy-4propylplienol
X
IE value,
15.23
X
X X
T = tentative. * Mass spectral data are reported in order of decreasing intensities with the molecular weight in italics.. = previous1:y reported in smoke condensate.
Figure 1. Separation of
hickory smoke tar by preparative gas chromatography. J. Agr. Food Chern., Vol. 22, No. 1,1974
125
VAN MEGEN
through a Watson-Biemann separator (Watson and Biemann, 1965) to the mass spectrometer, while the remainder of the effluent went to the flame ionization detector of the gas chromatograph. Retention indices (I, values) were determined by the method of van den Do01 and Kratz (1963) using standards of ethyl esters of n-aliphatic acids and programmed temperature glc. The IE values for the unknown compounds in the glc-ms runs were obtained by interpolation between peaks of unambiguously identified compounds of known IE values. For this reason and because of the influence of other components on retention times, the I E values for some of the later eluting peaks could not be determined precisely as for the earlier eluting ones. The I E values obtained on the open tubular columns were, in most cases, very close to those obtained on standard packed columns. RESULTS AND DISCUSSION
The compounds identified from all the trappings and methylated acid fraction are listed in Table I. Identification was made by matching the mass spectra of the unknown with reference spectra and verified by comparing the retention index values of the unknown compounds in the glc-ms runs to IE values for known compounds. Tentative identifications are indicated where the mass spectra were weak or mixed, or reference compounds were not available for comparison. Many compounds that were used for reference were commercially purchased samples. In some cases, the chemicals were synthesized by the authors. When sufficient material could be trapped out, nmr and ir spectra were obtained to aid in the identification. Many of the major constituents of the tar of hickory wood smoke have been identified. Although there has been no previous work on the tar of hickory wood smoke, several of these compounds have been previously reported in smoke condensate and extracts. This is noted in Table I. There were also several compounds found in smoke condensate which were not found in the tar. A possible reason is that the smoke condensate was obtained in a laboratory under controlled conditions while the material for this investigation was a commercially purchased sample of which little is known.
SUMMARY
As a result of this work on the tar of hickory wood smoke, we have identified 81 of the major constituents. Although many of the identified constituents can be considered to contribute to the hickory smoke aroma, it is not possible on the basis of this study to pinpoint one or more chemicals which could be described definitely as hickory smoke. ACKNOWLEDGMENT
The authors thank Anne Sanderson for her help in the mass spectral interpretation and Mort Jacobs for his assistance in obtaining and interpreting the nmr spectra. LITERATURE CITED Fiddler, W., Doerr, R. C., Wasserman, A. E., Salary, J. M., J. Aer. Food Chem. 14.659 (1966). Fiddler, W., Parker, ’W. E., Wasserman, A. E., Doerr, R. C., J. Agr. Food Chem. 15,757 (1967). Fiddler W., Doerr, R. C., Wasserman, A. E., J. Agr. Food Chem. 18, 310 (1970a).
Fiddler, W., Wasserman. A. E.. Doerr. R. C.. J. Apr. - Food Chem. 18,934 (1970b).
Foster, W. W., Simpson, T . H., J. Sci. FoodAgr. 12,363 (1961). Foster, W. W., Simpson, T. H., Campbell, D., J. Sci. Food Agr. 12,635 (1961).
Hamid, H. A., Saffle, R. L., J . Food Sci. 30,69 (1965). Hoff, J. E., Kapsalopoulou, A. J., J. Gas Chromatogr. 2, 296 (1964).
Jahnsen, V. J., Ph.D. Dissertation, Purdue University, Lafayette, Ind., 1961. Kornreich, M. R., Issenberg, P., J . Agr. Food C h e n . 20, 1109 (1972).
Love, S., Bratzler, L. J., J. Food Sci. 31,218 (1966). Lustre, A. O., Issenberg, P., J. Agr. Food Chem. 17,1387 (1969). Porter, R. W., Bratzler, L. J., Pearson, A . M., J. Food Sci. 30, 615 ( 1964).
Tilgner, D., Ziemba, A., Dju-Hak-Dju, “Symposium,on Advances in the Engineering of the Smoke Curing Process, Tilgner, D., Ed., Yugoslav Institute of Meat Technology, Beograd, Yugoslavia, 1962, p 18. van den Dool, H., Kratz, P. D., J . Chromatogr. 11,463 (1963). Watson, J. T., Biemann, K., Anal. Chem. 37,844 (1965). Ziemba, A., Prezenysl Spozywczy 2, 200; data reproduced by Draudt, H. N., Food Technol. 17,1557 (1963). Received for review May 7, 1973. Accepted October 31,1973.
Solubility Behavior of Soybean Globulins as a Function of pH and Ionic Strength Wim H. van Megen
The solubility behavior of partially purified soybean globulins as a function of pH and ionic strength of the dispersing medium was investigated. It appeared that, even at the isoelectric point, the soy protein could be dissolved easily up to very high concentrations, provided that the ionic strength of the solution exceeded a critical value which, a t p H 4.5, was about 0.7 for NaCl and NaZS04 and 0.25 for CaC12. Below the critical ionic strength a two-phase system was formed, consisting of a protein-poor upper layer
The majority of the soy proteins are insoluble a t their isoelectric point. Their solubility in dilute aqueous solution increases as the pH diverges from this value or, a t constant pH, as the salt concentration is increased (saltUnilever Research, Duiven, The Netherlands. 126
J. Agr. Food Chern., Vol. 22, No. 1,1974
and a viscous protein-rich lower layer. At pH 7.0 no phase separation was observed at very low ionic strength but, as the salt concentration was increased, a region was passed in which the solution demixed. Outside the regions of immiscibility only homogeneous systems were obtained. The composition of the protein-poor layers in the two-phase systems was in agreement with the well known protein extractability curves for soybean meal.
ing-in). Therefore these proteins have been classified as globulins. The typical solubility behavior of globulins has been clearly demonstrated on /?-lactoglobulin (Gronwall, 1942). It is also well known that salting-in of proteins may be followed by salting-out when the ionic strength is in-
